even if sustained nuclear fusion were possible, commericially viable?

ensabah6

even if sustained nuclear fusion via deuterium-tritium type reactions were possible, and the energy that is released exceeds the energy that is put in, and this energy is harnessed as a conventional water-steam generator, would such a design be commericially viable, based on the current cost of experimental fusion reactors? If this technology can be done for say 2 billion (the LHC was around 6 billion I understand) making use of superconducting magnets in a torus, would such an expensive technology ever pay itself off, in comparison to fossil fuel burning plants?

mgb_phys

The experimental fusion reactors are expensive because they are experimental.
The core of a fusion reactor should be cheaper and easier to build than a fission reactor. Most of the engineering, steam+generator set distribution grid etc, is the same for any station. Then if your fuel is free (at least for D-D reactors).

The biggest cost is probably still the lawyer's bills for the public enquiry .

Xnn

Fusion reactors actually consume more energy than they produce.
So, even if they could be constructed for free, and the fuel was free and there was no maintenace expenses and operators all worked for free, they would still not make economic sense.

Also, the cost of extracting deuterium from water is not simple and actually very expensive.

I'll wager a wild guess, that unless fusion reactors produce at least 10 times as much energy as they consume, they will not be economic.

vanesch

Your wild guess is about right ! People estimate that the Q-factor, which is essentially the ratio of the energy usefully extracted over the energy used to heat the plasma, must be around 15-20 for a commercial plant. The highest value reached as of now by JET is 0.7 if my memory is right. In order for this to happen, one needs a self-igniting plasma, that is, one needs a plasma that when it is heated initially, produces enough self-heating to sustain the fusion, and then one needs to keep this going on long enough.

QuantumPion

Ignition means Q is infinite because you are putting in zero energy and getting out megawatts. This is not necessarily required for commercial applications. As you said, a Q of 20 is what they are aiming for (put in 100 megawatts of heat and get out 2000).

mgb_phys

I wonder what the initial input is for a coal fired station?
Having tried to light a coal fire with matches->paper->sticks->coal do they cheat and just use a big bottle of barbecue lighter fluid?

gmax137

The burning coal in a generating station has been ground into dust and is entrained with compressed air. Toss in a match and off you go. Or, I guess you could think of the "Q" in terms of the power from the grid to start the pulverizers and fans. Or the diesel-gens you light off to do that if you're doing a black start...

mheslep

I think you are wrong on this.
1. The cost of tokamaks is intrinsic to the design. Once the design requires a neutron blanket, and D-D / D-T tokamaks do, they lose, as they must be much larger than a LWR fission reactor core of comparable power.
2. It appears Tokamaks can only come in one size - large - well above 1000MWe. Commercial viability requires financial viability; a large tokamak fusion plant is still going to require too much money up front and too much time to build.

Relevant comments from LANL fusion researcher Rick Nebel:
http://www.talk-polywell.org/bb/view...ighlight=#9834
and
http://www.talk-polywell.org/bb/view...ighlight=#9974

mgb_phys

I'm not sure power density matters. The power density of a field of wind turbines is very low.
I was basing it on the inside of a tokamak being an empty metal tube. You don't have the complex fuel structure and hot primary cooling circuit.
All the expensive complex stuff is outside the tokamak where it doesn't get so much of a dose so doesn't need special materials or decommisioning. The inside of the chamber does but you just fill it with concrete when you are done.

That's about the size you would want to build anyway, smaller than that and it's not worth the cost of the generator sets. Typically you have 2x680Mw reactors in a station, a coal station is usally 3-4Gw.

His other point about the cost doesn't really add up. The first atom bombs, computers, jet aircraft, rockets all needed a war time national project to build but they are now a little easier. It is probably more commercialisable than LANL's laser initiated pellet alternative.

mheslep

Per Nebel, he's measuring commercial viability by means of density of power per reactor mass, or more generally power per 'stuff' that costs money. There is no plant 'stuff' between wind turbines.

Yes I follow but I don't think that's a useful representation of proposed Tokamak designs. First, the metal tube is not really the reaction container, those expensive superconducting magnets are (rather the field created by them, and the circulation of the plasma). Then, the lithium blanket must be considered part of the reactor, as nuclear reactions are also occurring there. It is needed both to capture the power of the neutron flux, AND to generate the tritium fuel. So the fuel process is complex, if ~ self contained. Finally, that first wall or metal tube must be replaced frequently (again because of neutron flux), every year for the current design.

The point is industry generally avoids building big - its risky. So if 'very large' is indeed the only Tokamak size, then we are indeed tossing out it's commercial chances and relegating Tokamak fusion to government funding. That might be ok, but the OT is commercially viable fusion.
? The median coal plant in the US is 200-300MW; two hundred of the total are less than 10MWe. There's exactly one 4GWe coal plant in North America. Combined cycle gas turbines are sized even smaller, and they're the sweet spot for the industry.
http://www.sourcewatch.org/index.php...e-EIA_existing
http://www.gepower.com/prod_serv/pro...c/en/index.htm
On the low end we have ~1MW wind turbines, which obviously include the generator.

mgb_phys

I suppose in a big country long distance transmission lines are expensive.
A big chunk of the UK's power comes from 3 coal fired stations near where I used to live - they are 2.5-4Gw (coal works best with big stations) most of the rest comes form paired reactors or 1500Mw gas fired stations.

The real achievement of ITER is probably getting that many countries to agree on building something!

Morbius

Xnn,

There's NO REASON that we can't ultimately extract energy from fusion - after all that's
what Hydrogen bombs do - you get more FUSION energy from Hydrogen bombs than you put in.

However, we can't control the reaction at present to output a manageable amount of energy. But
that will change in the future. The National Ignition Facility at Lawrence Livermore National Laboratory
is expected to reach fusion "ignition" - more energy out than went in:

https://lasers.llnl.gov/

Actually, extraction of deutrerium is NOT prohibitively expensive - Canada separates heavy water
from light water ALL THE TIME for their CANDU reactors.

Dr. Gregory Greenman
Physicist

Morbius

mheslep,

Tokamaks aren't the only fusion reactor design. Lawrence Livermore National Laboratory is
working on a concept called LIFE:

https://lasers.llnl.gov/missions/ene...e_future/life/

https://lasers.llnl.gov/missions/ene...life_works.php

Dr. Gregory Greenman
Physicist

JeffKoch

The answer depends on the cost of the alternatives. If fossil fuels go the way of the creatures that supposedly made them, then there aren't many alternatives - you're looking at electric transportation and home heating and lighting, powered by nuclear reactors. Fission reactors are a near-term solution, fusion/fission hybrids are a long-term solution that requires a lot of R&D. If I knew I'd be alive to collect the money, I'd bet a fortune that in 100-200 years the world economy will be driven by fusion energy.

mheslep

Why aren't future fission reactors - gen '5' or so - a long term solution in your thinking? Thorium, fast breeders - all of that together appear to handle fuel and waste problems for a 1000 years. No doubt fusion will have net power in your time frame, but will they pay better than fission technology at the time? You just might lose your money.

JeffKoch

Neither of us will be around to find out, so both are safe bets. I suspect what will ultimately decide it is the cost of the fuel and dealing with the waste. Fission/fusion reactors will require a staggering expense to develop, but the cost is non-recurring - once you've figured it out, you build plants and make energy. Economical DD fusion is hard to imagine right now, but the fusion fuel is close to free and serves to burn up the fission byproducts - if I extend my timeframe far enough and assume our clever descendants will work out the details, it's hard to see why this isn't the winning bet. I could be wrong.